The Value of MR Imaging in Comparative Analysis of Spinal Infection in Adults: Pyogenic Versus Tuberculous

A diagnostic model for differentiating tuberculous spondylodiscitis from pyogenic spondylodiscitis based on pathogen-confirmed patients

OBJECTIVE

This study aimed to distinguish tuberculous spondylodiscitis (TS) from pyogenic spondylodiscitis (PS) based on laboratory, magnetic resonance imaging (MRI) and computed tomography (CT) findings. Further, a novel diagnostic model for differential diagnosis was developed.

METHODS

We obtained MRI, CT and laboratory data from TS and PS patients. Predictive models were built using binary logistic regression analysis. The receiver operating characteristic curve was analyzed. Both internal and external validation was performed.

RESULTS

A total of 81 patients with PS (n = 46) or TS (n = 35) were enrolled. All patients had etiological evidence from the focal lesion. Disc signal or height preservation, skip lesion or multi segment (involved segments ≥ 3) involvement, paravertebral calcification, massive sequestra formation, subligamentous bone destruction, bone erosion with osteosclerotic margin, higher White Blood Cell Count (WBC) and positive result of tuberculosis infection T cell spot test (T-SPOT.TB) were more prevalent in the TS group. A diagnostic model was developed and included four predictors: WBC<7.265 * (10^9/L), skip lesion or involved segments ≥ 3, massive sequestra formation and subligamentous bone destruction. The model showed good sensitivity, specificity, and total accuracy (91.4%, 95.7%, and 93.8%, respectively); the area under the receiver operating characteristic curve (AUC) was 0.981, similar to the results of internal validation using bootstrap resampling (1000 replicates) and external validation set, indicating good clinical predictive ability.

CONCLUSIONS

This study develop a good diagnostic model based on both CT and MRI, as well as laboratory findings, which may help clinicians distinguish between TS and PS.

A Comparative Factor Analysis and New Magnetic Resonance Imaging Scoring System for Differentiating Pyogenic Versus Tuberculous Spondylodiscitis

OBJECTIVE

This study aimed to compare and analyze differences in clinical and magnetic resonance imaging (MRI) findings between tuberculous spondylodiscitis (TbS) and pyogenic spondylodiscitis (PyS), and to develop and validate a simplified multiparameter MRIbased scoring system for differentiating TbS from PyS.

METHODS

We compared predisposing factors in 190 patients: 123 with TbS and 67 with PyS, confirmed by laboratory tests, culture, or pathology. Data encompassing patient demographics, clinical characteristics, laboratory results, and MRI findings were collected between 2015 and 2020. Data were analyzed using logistic regression methods, and selected coefficients were transformed into an MRI-based scoring system. Internal validation was performed using bootstrapping method.

RESULTS

Univariate analysis revealed that the significant risk factors associated with TbS included thoracic lesions, vertebral destruction > 50%, intraosseous abscess, thin-walled abscess, well-defined paravertebral abscess, subligamentous spreading, and epidural abscess. Multivariate analysis revealed that only thoracic lesions, absence of epidural phlegmon, subligamentous spreading, intraosseous abscesses, well-defined paravertebral abscesses, epidural abscesses, and absence of facet joint arthritis were independent predictive factors for TbS (all p < 0.05). These potential predictors were used to derive an MRI scoring system. Total scores ≥ 14/29 points significantly predicted the probability of TbS, with a sensitivity of 97.58%, specificity of 92.54%, and an area under the curve of 0.96 (95% confidence interval, 125.40-3,257.95).

CONCLUSION

This simplified MRI-based scoring system for differentiating TbS from PyS helps guide appropriate treatment when the causative organism is not identified.

Imaging of Spondylodiscitis: A Comprehensive Updated Review-Multimodality Imaging Findings, Differential Diagnosis, and Specific Microorganisms Detection

Spondylodiscitis is defined by infectious conditions involving the vertebral column. The incidence of the disease has constantly increased over the last decades. Imaging plays a key role in each phase of the disease. Indeed, radiological tools are fundamental in (i) the initial diagnostic recognition of spondylodiscitis, (ii) the differentiation against inflammatory, degenerative, or calcific etiologies, (iii) the disease staging, as well as (iv) to provide clues to orient towards the microorganisms involved. This latter aim can be achieved with a mini-invasive procedure (e.g., CT-guided biopsy) or can be non-invasively supposed by the analysis of the CT, positron emission tomography (PET) CT, or MRI features displayed. Hence, this comprehensive review aims to summarize all the multimodality imaging features of spondylodiscitis. This, with the goal of serving as a reference for Physicians (infectious disease specialists, spine surgeons, radiologists) involved in the care of these patients. Nonetheless, this review article may offer starting points for future research articles.

Magnetic resonance imaging features for differentiating tuberculous from pyogenic spondylitis: a meta-analysis

OBJECTIVE

To perform a meta-analysis comparing the MRI features of tuberculous and pyogenic spondylitis, using histopathological results and/or blood culture as the standard reference.

MATERIALS AND METHODS

PubMed, Embase, Web of Science, and Cochrane Library were searched for English-language studies on the MRI features of tuberculous and pyogenic spondylitis published between January 2010 and February 2023. Risk for bias and concerns regarding applicability were assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 tool. Pooled MRI features' proportions were calculated using a bivariate random-effects model.

RESULTS

Thirty-two studies met the inclusion criteria: 21 for tuberculous spondylitis, three for pyogenic spondylitis, and eight for both. Of the nine informative MRI features comparing tuberculous spondylitis to pyogenic spondylitis, involvement of ≥ 2 vertebral bodies (92% vs. 88%, P = .004), epidural extension (77% vs. 25%, P < .001), paravertebral collection (91% vs. 84%, P < .001), subligamentous spread (93% vs. 24%, P < .001), thin and regular abscess wall (94% vs. 18%, P < .001), vertebral collapse (68% vs. 24%, P < .001), and kyphosis (39% vs. 3%, P < .01) were more suggestive of tuberculous spondylitis, while disc signal change (82% vs. 95%, P < .001) and disc height loss (22% vs. 59%, P < .001) were more suggestive of pyogenic spondylitis.

CONCLUSION

Involvement of ≥ 2 vertebral vertebral bodies, soft tissue attribution, thin and regular abscess wall, vertebral collapse, and kyphosis were MRI features more common in tuberculous spondylitis, while disc signal change and height loss were more common in pyogenic spondylitis.

Development of a Diagnostic Model for Differentiating Tuberculous Spondylitis and Pyogenic Spondylitis With MRI: A Multicenter Retrospective Observational Study

STUDY DESIGN

Multicenter retrospective observational study.

OBJECTIVE

This study aimed to distinguish tuberculous spondylitis (TS) from pyogenic spondylitis (PS) using magnetic resonance imaging (MRI). Further, a novel diagnostic model for differential diagnosis was developed.

SUMMARY OF BACKGROUND DATA

TS and PS are the two most common spinal infections. Distinguishing between these types clinically is challenging. Delayed diagnosis can lead to deficits or kyphosis. Currently, there is a lack of radiology-based diagnostic models for TS and PS.

METHODS

We obtained radiologic images from MRI imaging of patients with TS and PS and applied the least absolute shrinkage and selection operator regression to select the optimal features for a predictive model. Predictive models were built using multiple logistic regression analysis. Clinical utility was determined using decision curve analysis, and internal validation was performed using bootstrap resampling.

RESULTS

A total of 201 patients with TS (n=105) or PS (n=96) were enrolled. We identified significant differences in MRI features between both groups. We found that noncontiguous multivertebral and single-vertebral body involvement were common in TS and PS, respectively. Vertebral bone lesions were more severe in the TS group than in the PS group (Z=-4.553, P <0.001). The patients in the TS group were also more prone to vertebral intraosseous, epidural, and paraspinal abscesses ( P <0.001). A total of 8 predictors were included in the diagnostic model. Analysis of the calibration curve and area under the receiver operating characteristic curve suggested that the model was well-calibrated with high prediction accuracy.

CONCLUSIONS

This is the largest study comparing MRI features in TS and PS and the first to develop an MRI-based nomogram, which may help clinicians distinguish between TS and PS.

Value of [18F]FDG PET/CT in Diagnosis and Management of Spondylodiscitis

Vertebral osteomyelitis is the third most common form of osteomyelitis in patients over 50 years of age.Whereas prompt (pathogen-directed) therapy is crucially associated with better outcomes, the heterogeneous clinical presentation of disease with unspecific symptoms often delays adequate treatment initiation. Diagnosis requires a careful investigation of medical history, clinical findings and diagnostic imaging, including magnetic resonance imaging and nuclear medicine techniques.Due to its high sensitivity, [18F]FDG PET/CT is becoming increasingly important in diagnosis and management of spondylodiscitis, especially in the postoperative setting with presence of spinal hardware or other implantable devices in which MRI is limited.

A diagnostic model for differentiating tuberculous spondylitis from pyogenic spondylitis: a retrospective case-control study

The purpose of this study was to describe and compare the clinical data, laboratory examination and imaging examination of tuberculous spondylitis (TS) and pyogenic spondylitis (PS), and to provide ideas for diagnosis and treatment intervention. The patients with TS or PS diagnosed by pathology who first occurred in our hospital from September 2018 to November 2021 were studied retrospectively. The clinical data, laboratory results and imaging findings of the two groups were analyzed and compared. The diagnostic model was constructed by binary logistic regression. In addition, an external validation group was used to verify the effectiveness of the diagnostic model. A total of 112 patients were included, including 65 cases of TS with an average age of 49 ± 15 years, 47 cases of PS with an average of 56 ± 10 years. The PS group had a significantly older age than the TS group (P = 0.005). In laboratory examination, there were significant differences in WBC, neutrophil (N), lymphocyte (L), ESR, CRP, fibrinogen (FIB), serum albumin (A) and sodium (Na). The difference was also statistically significant in the comparison of imaging examinations at epidural abscesses, paravertebral abscesses, spinal cord compression, involvement of cervical, lumbar and thoracic vertebrae. This study constructed a diagnostic model, which was Y (value of TS > 0.5, value of PS < 0.5) = 1.251 * X1 (thoracic vertebrae involved = 1, thoracic vertebrae uninvolved = 0) + 2.021 * X2 (paravertebral abscesses = 1, no paravertebral abscess = 0) + 2.432 * X3 (spinal cord compression = 1, no spinal cord compression = 0) + 0.18 * X4 (value of serum A)-4.209 * X5 (cervical vertebrae involved = 1, cervical vertebrae uninvolved = 0)-0.02 * X6 (value of ESR)-0.806 * X7 (value of FIB)-3.36. Furthermore, the diagnostic model was validated using an external validation group, indicating a certain value in diagnosing TS and PS. This study puts forward a diagnostic model for the diagnosis of TS and PS in spinal infection for the first time, which has potential guiding value in the diagnosis of them and provides a certain reference for clinical work.

Bilateral Percutaneous Transpedicular Drainage under Local Anesthesia for Thoracic Tuberculous Spondylitis

Tuberculous spondylitis is a common spinal infection. If surgical intervention is necessary, anterior debridement and anterior fixation are typically performed. However, a minimally invasive surgical strategy under local anesthesia seems rarely implemented. A 68-year-old man presented with severe pain in the left flank. Whole spinal magnetic resonance imaging revealed abnormal intensity of vertebral bodies from T6-9. A bilateral paravertebral abscess extending from T4-10 was suspected. The T7/T8 intervertebral disc was destroyed, but severe vertebral deformity or spinal cord compression was not observed. Bilateral percutaneous transpedicular drainage under local anesthesia was planned. The patient was set in the prone position. Under the guide of a biplanar angiographic system, the bilateral drainage tubes were placed paravertebrally in the abscess cavity. The left flank pain improved after the procedure. Laboratory culture of the pus specimen confirmed a diagnosis of tuberculosis. A chemotherapy regimen for tuberculosis was soon initiated. The patient was discharged during postoperative week 2, with continuation of chemotherapy for tuberculosis. Percutaneous transpedicular drainage under local anesthesia can be effective in the management of thoracic tuberculous spondylitis without severe vertebral deformity or compression of the spinal cord by an abscess.

Value of [18F]FDG PET/CT in diagnosis and management of spondylodiscitis

Vertebral osteomyelitis is the third most common form of osteomyelitis in patients over 50 years of age.Whereas prompt (pathogen-directed) therapy is crucially associated with better outcomes, the heterogeneous clinical presentation of disease with unspecific symptoms often delays adequate treatment initiation. Diagnosis requires a careful investigation of medical history, clinical findings and diagnostic imaging, including magnetic resonance imaging and nuclear medicine techniques.Due to its high sensitivity, [¹⁸F]FDG PET/CT is becoming increasingly important in diagnosis and management of spondylodiscitis, especially in the postoperative setting with presence of spinal hardware or other implantable devices in which MRI is limited.

The Incidence, Changes and Treatments of Cervical Deformity After Infection and Inflammation

A healthy cervical spine with normal movement is the basis of many daily activities and is essential for maintaining a good quality of life. However, the alignment, fusion, and structure of the cervical spine can change for various reasons, leading to cervical deformity, mainly kyphosis. Approximately 5%‒20% of spinal infections in the cervical spine cause cervical deformity. The deformity can recover early; however, the disease's long-term existence or the continuous action of abnormal stress may lead to intervertebral fusion and abnormal osteophytes. Many gaps and controversies exist regarding infectious cervical deformities, including a lack of clear definitions and an acceptable classification system thereby requiring further research. Moreover, there is no consensus on the indications for postinfectious cervical deformity associated with <i>Mycobacterium tuberculosis</i>, <i>Staphylococcus aureus</i>, and Brucellosis. Therefore, we reviewed and discussed the incidence, clinical manifestations, changes, and treatment of infectious and inflammatory secondary cervical deformities from common to rare to provide a theoretical basis for clinical decision-making.

Spondylodiscitis Presenting to a Chiropractor: A Case Report and Literature Review

An 80-year-old man under combination therapy for pulmonary tuberculosis presented to a chiropractor with a one-month history of worsening chronic low back pain, yet denied having any respiratory symptoms, weight loss, or night sweats. Two weeks prior, he saw an orthopedist who ordered lumbar radiographs and magnetic resonance imaging (MRI), showing degenerative changes and subtle findings of spondylodiscitis, but was treated conservatively with a nonsteroidal anti-inflammatory drug. The patient was afebrile, yet considering his older age and worsening symptoms, the chiropractor ordered a repeat MRI with contrast, which revealed more advanced findings of spondylodiscitis, psoas abscesses, and epidural phlegmon, and referred the patient to the emergency department. A biopsy and culture confirmed Staphylococcus aureus infection and were negative for Mycobacterium tuberculosis. The patient was admitted and treated with intravenous antibiotics. We conducted a literature review revealing nine previously published cases of patients with spinal infection presenting to a chiropractor, who were typically afebrile men with severe low back pain. Chiropractors rarely encounter patients with undiagnosed spinal infections and should manage those suspected of infection with urgency via advanced imaging and/or referral.

A Predictive Clinical-Radiomics Nomogram for Differentiating Tuberculous Spondylitis from Pyogenic Spondylitis Using CT and Clinical Risk Factors

OBJECTIVE

The study aimed to develop and validate a nomogram model with clinical risk factors and radiomic features for differentiating tuberculous spondylitis (TS) from pyogenic spondylitis (PS).

METHODS

A total of 254 patients with TS (n = 141) or PS (n = 113) were randomly divided into training (n = 180) and validation (n = 74) groups. In addition, 43 patients (TS = 22 and PS = 21) were collected to construct a test cohort. t-test analysis, de-redundancy analysis, and minimum absolute shrinkage and selection operator (lasso) algorithm were utilized on the training set to obtain the optimal radiomics features from computed tomography (CT) for constructing the radiomics model and determine the radiomics score (Rad-score). Eight clinical risk predictors were identified to develop the clinical model. Combined with clinical risk predictors and Rad-scores, a nomogram model was constructed using multivariate logistic regression analysis.

RESULTS

A total of 1781 features were extracted, and 12 optimal radiomic features were utilized to construct the radiomic model and determine the Rad-score. The combined clinical radiomics model revealed good discrimination performance in both the training cohort and the validation cohort (AUC = 0.891 and 0.830) and was superior to the clinical (AUC = 0.807 and 0.785) and radiomics (AUC = 0.796 and 0.811) models. The calibration curve and DCA also depicted that the nomogram had better clinical efficacy. The discriminative performance of the model is well validated in the test cohort (AUC=0.877).

CONCLUSION

The clinical radiomic nomogram could serve as a promising predictive tool for differentiating TS from PS, which could be helpful for clinical decision-making.

Computer-Aided Diagnosis of Spinal Tuberculosis From CT Images Based on Deep Learning With Multimodal Feature Fusion

Background

Spinal tuberculosis (TB) has the highest incidence in remote plateau areas, particularly in Tibet, China, due to inadequate local healthcare services, which not only facilitates the transmission of TB bacteria but also increases the burden on grassroots hospitals. Computer-aided diagnosis (CAD) is urgently required to improve the efficiency of clinical diagnosis of TB using computed tomography (CT) images. However, classical machine learning with handcrafted features generally has low accuracy, and deep learning with self-extracting features relies heavily on the size of medical datasets. Therefore, CAD, which effectively fuses multimodal features, is an alternative solution for spinal TB detection.

Methods

A new deep learning method is proposed that fuses four elaborate image features, specifically three handcrafted features and one convolutional neural network (CNN) feature. Spinal TB CT images were collected from 197 patients with spinal TB, from 2013 to 2020, in the People’s Hospital of Tibet Autonomous Region, China; 3,000 effective lumbar spine CT images were randomly screened to our dataset, from which two sets of 1,500 images each were classified as tuberculosis (positive) and health (negative). In addition, virtual data augmentation is proposed to enlarge the handcrafted features of the TB dataset. Essentially, the proposed multimodal feature fusion CNN consists of four main sections: matching network, backbone (ResNet-18/50, VGG-11/16, DenseNet-121/161), fallen network, and gated information fusion network. Detailed performance analyses were conducted based on the multimodal features, proposed augmentation, model stability, and model-focused heatmap.

Results

Experimental results showed that the proposed model with VGG-11 and virtual data augmentation exhibited optimal performance in terms of accuracy, specificity, sensitivity, and area under curve. In addition, an inverse relationship existed between the model size and test accuracy. The model-focused heatmap also shifted from the irrelevant region to the bone destruction caused by TB.

Conclusion

The proposed augmentation effectively simulated the real data distribution in the feature space. More importantly, all the evaluation metrics and analyses demonstrated that the proposed deep learning model exhibits efficient feature fusion for multimodal features. Our study provides a profound insight into the preliminary auxiliary diagnosis of spinal TB from CT images applicable to the Tibetan area.

ACR Appropriateness Criteria® Suspected Spine Infection

Spine infection is both a clinical and diagnostic imaging challenge due to its relatively indolent and nonspecific clinical presentation. The diagnosis of spine infection is based upon a combination of clinical suspicion, imaging evaluation and, when possible, microbiologic confirmation performed from blood cultures or image-guided percutaneous or open spine biopsy. With respect to the imaging evaluation of suspected spine infection, MRI without and with contrast of the affected spine segment is the initial diagnostic test of choice. As noncontrast MRI of the spine is often used in the evaluation of back or neck pain not responding to conservative medical management, it may show findings that are suggestive of infection, hence this procedure may also be considered in the evaluation of suspected spine infection. Nuclear medicine studies, including skeletal scintigraphy, gallium scan, and FDG-PET/CT, may be helpful in equivocal or select cases. Similarly, radiography and CT may be appropriate for assessing overall spinal stability, spine alignment, osseous integrity and, when present, the status of spine instrumentation or spine implants. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment.

MRI in differential diagnosis between tuberculous and pyogenic spondylodiscitis

PURPOSE

The aim of the study was to investigate whether MRI findings together with epidemiological data could help in differentiating between tuberculous and pyogenic spondylodiscitis.

METHODS

Clinical records of 260 patients with a suspicion of spondylodiscitis were analysed. Patients were selected using the following inclusion criteria: confirmed diagnosis of spondylodiscitis either from pyogenic bacteria or from Mycobacterium tuberculosis and contrast-enhanced MRI performed before treatment. Clinical data concerning age, sex and country-of-origin were also collected. For each patient, several MRI-features were evaluated by two-expert musculoskeletal radiologists. A chi-squared test and a multiple logistic regression were used to find the best predictors of tuberculous or pyogenic spondylodiscitis.

RESULTS

114 patients were retrospectively enrolled, 30 with tuberculous and 84 with pyogenic spondylodiscitis. We found 18 MRI-features, significantly different between the two groups. Among these, the most strongly associated with tuberculous spondylodiscitis were: heterogeneous vertebral signal on T1w-sequences (Odds Ratio(OR) = 205.759-p < 0.001), presence of epidural abscess (OR = 86.221-p < 0.001), severe vertebral destruction (OR = 10.017-p < 0.001) and absence of epidural phlegmon (OR = 86.221-p < 0.001). Moreover, patients coming from countries with a middle-high prevalence of tuberculosis were more frequently affected by tuberculous spondylodiscitis than others were (OR = 229.136-p < 0.001). The best prediction model demonstrated a correct classification rate of 94.7%.

CONCLUSION

To the best of our knowledge this is the largest study comparing MRI-features of tuberculous and pyogenic spondylodiscitis. The above-mentioned MRI-features and epidemiological data are crucial in the differential diagnosis between these two entities, guiding the choice of the appropriate therapy, especially when a pathogen cannot be clearly identified with other modalities.

Assessment of diffusion tensor imaging in differentiation between pyogenic and tuberculous spondylitis

PURPOSE

to assess diffusion tensor imaging (DTI); an emerging technique for differentiation between pyogenic and tuberculous spondylitis.

PATIENTS AND METHODS

The study was carried out on 33 patients with infective spondylitis performing conventional MRI and DTI. The mean diffusivity (MD) and fractional anisotropy (FA) of the affected vertebral body were calculated by two readers.

RESULTS

The MD of pyogenic spondylitis of both readers (1.48 ± 0.09 and 1.47 ± 0.08 × 10^-3 mm²/s) were significantly higher values (P = 0.001) than tuberculous spondylitis (1.11 ± 0.15 and 1.18 ± 0.08 × 10^-3 mm²/s). The FA of pyogenic spondylitis of both readers (0.18 ± 0.09 and 0.20 ± 0.08) were significantly lower values (P = 0.001) than tuberculous spondylitis (0.30 ± 0.05 and 0.32 ± 0.03). There was a strong inter-reader agreement between both readers using MD (K = 0.963) and FA (K = 0.858). The thresholds MD and FA used for differentiating pyogenic and tuberculous spondylitis of both readers were 1.37 and 1.33 × 10^-3 mm²/s and 0.21 and 0.25 with the area under the curve (AUC) of 0.927 and 0.831 respectively. Combined MD and FA revealed increased AUC to 0.97 and 0.98 of both readers respectively.

CONCLUSION

DTI with its parameters can be considered a noninvasive beneficial quantitative method that can help in differentiation between pyogenic and tuberculous spondylitis.

A diagnostic model for differentiating tuberculous spondylitis from pyogenic spondylitis on computed tomography images

OBJECTIVES

To develop and evaluate a logistics regression diagnostic model based on computer tomography (CT) features to differentiate tuberculous spondylitis (TS) from pyogenic spondylitis (PS).

METHODS

Demographic and clinical features were collected from the Electronic Medical Record System. Data of bony changes seen on CT images were compared between the PS (n = 61) and TS (n = 51) groups using the chi-squared test or t test. Based on features that were identified to be significant, a diagnostic model was developed from a derivation set (two thirds) and evaluated in a validation set (one third). The sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were calculated.

RESULTS

The width of bone formation around the vertebra and sequestrum was greater in the TS group. There were significant differences between the two groups in the horizontal and longitudinal location of erosion and the morphology of axial bone destruction and sagittal residual vertebra. Kyphotic deformity and overlapping vertebrae were more common in the TS group. A diagnostic model that included eight predictors was developed and simplified to include the following six predictors: width of the bone formation surrounding the vertebra, longitudinal location, axial-specific erosive morphology, specific morphology of the residual vertebra, kyphotic deformity, and overlapping vertebrae. The simplified model showed good sensitivity, specificity, and total accuracy (85.59%, 87.80%, and 86.50%, respectively); the AUC was 0.95, indicating good clinical predictive ability.

CONCLUSIONS

A diagnostic model based on bone destruction and formation seen on CT images can facilitate clinical differentiation of TS from PS.

KEY POINTS

• We have developed and validated a simple diagnostic model based on bone destruction and formation observed on CT images that can differentiate tuberculous spondylitis from pyogenic spondylitis. • The model includes six predictors: width of the bone formation surrounding the vertebra, longitudinal location, axial-specific erosive morphology, specific morphology of the residual vertebra, kyphotic deformity, and overlapping vertebrae. • The simplified model has good sensitivity, specificity, and total accuracy with a high AUC, indicating excellent predictive ability.